Subscriber access provided by Nanyang Technological Univ
Article
A combined LC-MS/MS and molecular networking approach reveals new cyanotoxins from the 2014 cyanobacterial bloom in Green Lake, Seattle Roberta Teta, Gerardo Della Sala, Evgenia Glukhov, Lena Gerwick, William H Gerwick, Alfonso Mangoni, and Valeria Costantino Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.5b04415 • Publication Date (Web): 14 Nov 2015 Downloaded from http://pubs.acs.org on November 17, 2015
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 26
Environmental Science & Technology
1
A combined LC-MS/MS and molecular networking approach
2
reveals new cyanotoxins from the 2014 cyanobacterial bloom in
3
Green Lake, Seattle
4
Roberta Teta,† Gerardo Della Sala,† Evgenia Glukhov,‡ Lena Gerwick,‡ William H. Gerwick,‡
5
Alfonso Mangoni,† Valeria Costantino†,*
6
†
Napoli Federico II, via D. Montesano 49, 80131 Napoli, Italy
7 8 9 10
The NeaNat Group (www.neanat.unina.it), Dipartimento di Farmacia, Università degli Studi di
‡
Center for Marine Biotechnology and Biomedicine, Scripps Institution of Oceanography, and Skaggs School of Pharmacy and Pharmaceutical Sciences, University of California at San Diego, 9500 Gilman Drive, MC 0212, La Jolla, CA 92093-0212, USA
11
ABSTRACT: Cyanotoxins obtained from a freshwater cyanobacterial collection at Green Lake,
12
Seattle during a cyanobacterial harmful algal bloom in the summer of 2014 were studied using a
13
new approach based on molecular networking analysis of liquid chromatography-tandem mass
14
spectrometry (LC-MS/MS) data. This MS-networking approach is particularly well suited for the
15
detection of new cyanotoxin variants, and resulted in the discovery of three new cyclic peptides,
16
namely microcystin-MhtyR (6) which comprised about half of the total microcystin content in the
17
bloom, and ferintoic acids C (12) and D (13). Structure elucidation of 6 was aided by a new
18
microscale methylation procedure. Metagenomic analysis of the bloom using the 16S-ITS rRNA
19
region identified Microcystis aeruginosa as the predominant cyanobacterium in the sample.
20
Fragments of the putative biosynthetic genes for the new cyanotoxins were also identified, and their
21
sequences correlated to the structure of the isolated cyanotoxins.
22
ACS Paragon Plus Environment
1
Environmental Science & Technology
Page 2 of 26
23
24
Graphics for TOC
25
ACS Paragon Plus Environment
2
Page 3 of 26
Environmental Science & Technology
26
INTRODUCTION
27
Cyanobacteria have a long history of ecological and health impacts. This unique class of
28
photosynthetic microorganisms can generally survive in nearly all phototrophic aquatic
29
environments, including recreational water bodies, fisheries, and reservoirs. In the last two decades,
30
worldwide attention has been given to the ecological effects of cyanobacteria and to their
31
production of secondary metabolites. They represent a not yet fully explored source of novel lead
32
compounds for drug discovery, such as the dolastatins,1 cryptophycins,2 and curacins,3 that have in
33
turn inspired the development of synthetic analogues with improved bioactivity and
34
pharmacokinetics. Some of these have successfully reached the phase II and phase III of clinical
35
trials, and one is a FDA clinically approved drug.4 In addition, it has been shown that symbiotic
36
cyanobacteria are often the real producers of secondary metabolites that were originally isolated and
37
thought to be produced by invertebrates.5,6,7 Interestingly, smenothiazole B,8 recently isolated from
38
the sponge Smenospongia aurea during our anticancer screening program, closely resembles that of
39
the neurotoxin jamaicamide B, 9 a hybrid peptide/polyketide obtained from the cyanobacterium
40
Moorea producens (formerly known as Lyngbya majuscula). The two molecules share a similar
41
vinyl chloride moiety and alkynyl terminus.
42
On the other hand, cyanobacteria growing in freshwater and marine recreational areas may
43
have a strongly negative impact on human health. Some species can form extensive blooms, and
44
there is evidence that these are increasing during recent decades due to nutrient enrichment,
45
especially phosphorus. 10 Many cyanobacteria produce toxic secondary metabolites (cyanotoxins)
46
with differing effects on health, ranging from mild skin irritations to severe illness.11 Economic
47
losses from freshwater and marine cyanobacterial blooms in the United States account for millions
48
of dollars per year, and include monitoring programs, closure of recreational areas and fisheries,
49
loss of livestock, water treatment and population wellness related costs.12
50
Because cyanobacteria are an attractive research topic from both the risks as well as potential
51
remedies they represent, we have initiated a worldwide program to study their chemistry and
ACS Paragon Plus Environment
3
Environmental Science & Technology
Page 4 of 26
52
biochemistry. Here we report our analysis of a sample of the cyanobacterial scum collected at Green
53
Lake, Seattle (WA, USA) during the 2014 summer bloom. This study involved a new approach that
54
combined liquid chromatography (LC)-high resolution mass spectrometry (HRMS) with automated
55
data analysis using the molecular networking technique,13 and led to the rapid identification of one
56
new microcystin (MC) (6) and two new ferintoic acid congeners (12 and 13). In addition,
57
metagenomic analysis of the sample allowed us to identify the bloom-forming cyanobacteria as a
58
strain of Microcystis aeruginosa, and to detect a fragment of the mcyBAd1 gene from the MC
59
biosynthetic cluster. The sequence and predicted function of this gene nicely correlate with the
60
structure of the MCs detected in the sample.
61
MATERIAL AND METHODS
62
Collection and Extraction: A sample of cyanobacterial scum was collected at Green Lake,
63
Seattle (WA, USA) on September 5, 2014 during a cyanobacterial harmful algal bloom. A voucher
64
sample is stored at Dipartimento di Farmacia, Università degli Studi di Napoli “Federico II” with
65
the reference number SEAGL14. An aliquot (2 mL) of the cyanobacterial suspension was sonicated
66
for 5 min and extracted with BuOH. The organic and aqueous layers were separated, filtered
67
(Whatman, 0.2 µm, PTFE) and dried, yielding 1 mg of dry organic extract; the extract was
68
resuspended in H2O at a concentration of 5 mg/mL and subsequently analyzed by LC-HRMS.
69
The BuOH extract was subjected to preparative reversed-phase HPLC separation on a Agilent
70
1260 Infinity Quaternary LC apparatus equipped with a Diode-Array Detector (DAD) [column 150
71
× 4.6 mm, 5 µm, Kinetex C18; eluent A: H2O; eluent B: ACN; gradient: 20→99% B, over 22 min,
72
flow rate 0.2 mL/min], which afforded two fractions (fraction A, tR = 15 min, and fraction B, tR =
73
16 min) containing, respectively, MC-HtyrR (5) and MC-MhtyR (6) (Figure 2) as the sole
74
microcystins, as determined by LC-HRMS analysis.
75
LC-HRMS and LC-HRMS/MS: Experiments were performed using a Thermo LTQ Orbitrap
76
XL high-resolution ESI mass spectrometer coupled to an Agilent model 1100 LC system, which
ACS Paragon Plus Environment
4
Page 5 of 26
Environmental Science & Technology
77
included a solvent reservoir, in-line degasser, binary pump, and refrigerated autosampler. A 5 µm
78
Kinetex C18 column (50 × 2.1 mm), maintained at 25 °C, was operated using a gradient elution of
79
H2O and ACN running at 200 µL/mim. The gradient program was as follows: 10% ACN for 5 min,
80
10%-99% ACN over 17 min, 99% ACN for 3 min. All the mass spectra were recorded in the
81
positive-ion mode. MS parameters were a spray voltage of 5 kV, a capillary temperature of 230 °C,
82
a sheath gas rate of 15 units N2 (ca. 150 mL/min), and an auxiliary gas rate of 5 units N2 (ca. 50
83
mL/min). Data were collected in the data-dependent acquisition (DDA) mode, in which the first and
84
second most intense ions of a full-scan mass spectrum were subjected to tandem mass spectrometry
85
(MS/MS) analysis. MS/MS scans were obtained for selected ions with CID fragmentation, isolation
86
width 2.0, normalized collision energy 36, Activation Q 0.250, and activation time 30 ms. Mass
87
data were analyzed using the Thermo Xcalibur software.
88
Methylation of microcystins. A mixture containing MC-LR and MC-YR (Sigma) (1.75 µg
89
each) in 350 µL of MeOH was treated with (trimethylsilyl)diazomethane (2M solution in Et2O, 100
90
µL).14 The reaction mixture was kept at room temperature overnight, then dried and redissolved in
91
H2O/MeOH (9:1) (100 µL) for subsequent LC-HRMS analysis. Similarly, fraction A (0.2 mg) and
92
fraction B (0.5 mg) from the HPLC separation described above were each dissolved in 400 µL of
93
MeOH and methylated with (trimethylsilyl)diazomethane (50 µL, 2M solution in Et2O) as described
94
above.
95
Molecular Networking: The mass spectral data were converted to mzXML format using the
96
program msconvert.15 A molecular network was obtained using the online workflow at the GNPS
97
website.16 The data was then clustered with MS-Cluster with a parent mass tolerance of 2.0 Da and
98
a MS/MS fragment ion tolerance of 0.1 Da to create consensus spectra. Further, consensus spectra
99
that contained less than 2 spectra were discarded. A network was then created where edges were
100
filtered to have a cosine score above 0.6 and more than 10 matched peaks. Furthermore, edges
101
between two nodes were kept in the network only if each of the nodes appeared in each other's
102
respective top 10 most similar nodes. The spectra in the network were then searched against
ACS Paragon Plus Environment
5
Environmental Science & Technology
Page 6 of 26
103
GNPS's spectral libraries. The data were then imported into Cytoscape 2.8.317 and displayed as a
104
network of nodes and edges. The network was organized with the FM3 layout plug-in. 18
105
Dereplication and search for analogues were performed separately after the creation of the network.
106
All matches within the library spectra were required to have a score above 0.7 and at least 10
107
matched peaks with a parent mass tolerance of 2.0 Da and a MS/MS fragment ion tolerance of 0.5
108
Da. Analogue search was enabled against the library with a maximum mass shift of 200 Da.
109
DNA isolation, PCR amplification, and clone library construction. Metagenomic DNA
110
from the Green Lake cyanobacteria was extracted from a pellet (about 40 mg) obtained by
111
centrifugation of the cyanobacterial scum. Primers used in the PCR experiments were the 16S-ITS
112
rRNA cyanobacterial specific primers CYA359F (5’-GGGGAATYTTCCGCAATGGG-3’) 19 and
113
373R
114
CTATGTTATTTATACATCAGG -3’) and RAA (5’- CTCAGCTTAACTTGATTATC -3’),21 and
115
the
116
CCNCGDATYTTNACYTG-3’).21 PCR products were subcloned with the TOPO TA cloning® kit
117
(Invitrogen) and sequenced. Sequences were analyzed by the RDP Seq Match tool,22 BLASTn23
118
and BLASTx. Full experimental details are reported in the Supporting Information.
119
(5’-CTAACCACCTGAGCTAAT-3’),
degenerate
primers
MTF
20
the
specific
mcyB
primers
(5’-GCNGGYGGYGCNTAYGTNCC-3’)
and
FAA
MTR
(5’-
(5’-
Partial sequences of 16S-ITS rDNA from Microcystis aeruginosa SEAGL14, mcyB_Ad1, and
120
ferB_Ad were deposited into GenBank under the accession numbers KT359577- KT359579.
121
RESULTS
122
Molecular Networking. A cyanobacterial bloom sample collected from Green Lake (Seattle,
123
WA, USA) during the bloom occurring in the summer 2014 was extracted and analyzed by LC-
124
HRMS and LC-HRMS/MS. Data obtained were used to generate a molecular network using the
125
Global Natural Product Social Molecular Network (GNPS)16 (See Experimental section for details).
126
The mass spectral molecular networking results in a fast identification of known metabolites from
127
natural extracts (dereplication)13 as well as new analogues. 24 Molecular networking provides a
ACS Paragon Plus Environment
6
Page 7 of 26
Environmental Science & Technology
128
visual representation of structural relationships as revealed by MS/MS data. A single chemical
129
species is represented as a node and the relatedness between compounds is represented by an edge.
130
Libraries containing a large number of MS/MS spectra of known natural products are available
131
at the GNPS website for the purpose of dereplication (additionally, users can provide their own
132
libraries). During the process of generation of the network, the spectra from one or more LC-
133
MS/MS runs are compared pairwise, and each spectrum is also compared with spectra in the
134
libraries. This strategy allows the identification not only of known metabolites, but also of their
135
structural analogues (either new or known compounds which are not present in the library) in a fast
136
and automatic way, and directly from raw LC-MS/MS data. To date, this new approach has not yet
137
been used in the analysis of a cyanobacterial bloom.
138
Because the data obtained in this study were high-resolution MS/MS spectra, we reasoned that
139
smaller values for the MS/MS fragment ion tolerance than the default 0.5 Da would improve the
140
quality of the network and minimize false clusters between structurally unrelated compounds. After
141
some trials, we empirically determined an optimal value of 0.1 Da for this parameter. Moreover,
142
because peptides are known to contain a large number of fragment ions, we selected a threshold of
143
10 matched peaks (compared to the default of 6 peaks) for two nodes to be connected. The Green
144
Lake two-dimensional network obtained using these parameters, and visualized using the Cytoscape
145
software,17 is displayed in Figure 1. The network contains 13 clusters ranging from 2 to 10 nodes.
146
Dereplication and search for analogues were performed separately after the creation of the
147
network, in part because not all of the available libraries of natural products contained high-
148
resolution spectra. Therefore, the default MS/MS fragment ion tolerance of 0.5 Da was restored,
149
and the specific dereplication workflow available at the GNPS website was used. This resulted in
150
the identification of three known cyanobacterial compounds, namely microcystins LR (1), 25
151
ferintoic acid A (11),26 and aerucyclamide A,27 along with several analogues of these metabolites.
152
The corresponding nodes of Figure 1 are color-coded accordingly.
ACS Paragon Plus Environment
7
Environmental Science & Technology
Page 8 of 26
153
While the molecular networking algorithm allows a fast and automated identification of
154
analogs of known compounds, it does not directly provide a chemical structure for these analogues.
155
Therefore, the structures of the analogues of microcystin and ferintoic acids were established from
156
an in-depth analysis of the high resolution MS/MS spectra combined with nanogram-scale chemical
157
derivatization.
158 159
Figure 1. Molecular network from Seattle Green Lake organic extract with a cosine similarity score cutoff of
160
0.60. Nodes are labeled with parent m/z ratio; edge thickness is related to cosine similarity score.
161
Microcystins. MCs are monocyclic heptapeptides biosynthesized via the nonribosomal peptide
162
synthetase (NRPS) pathway as encoded by the mcy gene cluster.28 MCs contain a common set of
163
five amino acids (including the unique β-amino acid 3-amino-9-methoxy-2,6,8-trimethyl-10-
164
phenyl-4,6-decadienoic acid, or Adda) and two variable L-amino acids at positions 2 and 4. MCs are
165
named after their variable L-amino acids. The most common MC, which contains leucine (L) at
166
position 2 and arginine (R) at position 4, is therefore called MC‐LR (1, Figure 2). Substitutions at
167
position 2 and 4 give rise to more than 20 primary MC analogues, with alterations in other
168
constituent amino acids resulting in more than 100 reported MCs to date.29,30 MCs are hepatotoxins;
169
they are strong inhibitors of type 1 and 2A serine protein phosphatases (PP1 and PP2A).31 These
ACS Paragon Plus Environment
8
Page 9 of 26
Environmental Science & Technology
170
enzymes are vital to various cellular processes such as cell growth and tumor suppression and
171
therefore these toxins are potent potential cancer promoters.32
172 173
Figure 2. (a) Structures of MC congeners found in Green Lake, Seattle. (b) The MC cluster from the Green
174
Lake extract. Adda is 3-amino-9-methoxy-2,6,8-trimethyl-10-phenyl-4,6-decadienoic acid; MeAsp is β-
175
methyl aspartic acid; Mdha is N‐methyldehydroalanine; (H4)Tyr is 1,2,3,4-tetrahydrotyrosine; Htyr is
176
homotyrosine; Mhtyr is N‐methylhomotyrosine.
177
The MC cluster contained six nodes, indicating the presence of five compounds closely related
178
to MC-LR (1, m/z 995.5535). The relative amounts of the six MCs were estimated from the areas of
179
the relevant peaks in the extracted-ion chromatograms from the LC-MS run (Figure 2). The
180
molecular weight of 2 (m/z 1029.5361) and 3 (m/z 1045.5320) suggested they were the relatively
181
common MC variants, MC-FR and MC-YR, respectively. This was confirmed by comparison of
182
their retention times and MS/MS spectra with those of authentic standards. The high-resolution
183
MS/MS fragmentation patterns of 1-3 were studied in detail (Figure 3). Most fragment ions derived
184
from the cleavage of two amide bonds, which is the typical fragmentation mode of cyclic
185
peptides.33 In all of these fragment ions, the positive charge was retained on the fragment containing
186
arginine. Fragments deriving from the loss of part of the Adda residue (ions a, l, and m in Figure 3)
187
were also observed. These analyses completed and confirmed previous interpretations of low-
ACS Paragon Plus Environment
9
Environmental Science & Technology
Page 10 of 26
188
resolution34 and high-resolution35 spectra of MCs, and provided a basis for the subsequent analysis
189
of the MS/MS spectra of MCs 4-6.
190 191
Figure 3. High-resolution MS/MS fragmentations of MCs. Fragments which are mass shifted and therefore
192
contain the variable amino acid residue are labeled in red, while the non-shifting fragments are labeled in
193
blue.
194
The molecular formula of 4 (m/z 1049.5633) indicated the presence of four additional hydrogen
195
atoms compared to MC-YR (3). Analysis of the MS/MS spectrum showed that all of the ions ACS Paragon Plus Environment
10
Page 11 of 26
Environmental Science & Technology
196
containing amino acid 2 were shifted to higher mass by 4 Da, while the remaining ions were
197
unchanged, thus locating the additional four H-atoms on amino acid 2 (Tyr in MC-YR). These data
198
are consistent with MC-(H4)YR, a known MC containing a 1,2,3,4-tetrahydrotyrosine residue.36,37
199
Compound 5 (m/z 1059.5480) showed an additional CH2 compared with MC-YR (3); this was
200
also located at the amino acid 2 position (Figure 3), suggesting the presence of a homotyrosine
201
residue. Homotyrosine is frequently found in MCs, but its presence (rather than the isomeric O-
202
methyl tyrosine that is also found in cyanobacterial peptides) 38 is not always confirmed
203
experimentally. In our case, the presence of a homotyrosine residue was confirmed by chemical
204
derivatization as described in the following section.
205
The new MC 6 (m/z 1073.5630) showed an additional CH2 compared to MC-HtyR (5), and
206
once again, MS/MS fragment ions demonstrated that the additional 14 Da mass belonged to amino
207
acid 2 (Figure 3). However, because all fragment ions (except those involving cleavage of Adda)
208
contained intact amino acids, it was not possible to establish the exact location of this CH2 group on
209
the homotyrosine residue from MS/MS data alone. Two structures seemed most reasonable, either
210
an O-methylated homotyrosine or an N-methylated homotyrosine residue at position 2. It was not
211
possible to obtain 6 in sufficient quantity for NMR analysis. Therefore, a microscale O-methylation
212
procedure was developed and used to distinguish between the two possible structures.
ACS Paragon Plus Environment
11
Environmental Science & Technology
Page 12 of 26
213
214
Figure 4. Structures of methylated MCs obtained by methylation of standard samples of MC-LR (1) and
215
MC-YR (3) and of MC-HtyR (5) and MC-MhtyR (6) from the Green Lake extract, and high-resolution
216
MS/MS fragmentations of the methylated standards which were used to localize the sites of methylation.
217
Fragments that are mass shifted and therefore contain the variable amino acid are labeled in red whereas the
218
non-shifting fragments are labeled in blue. Fragment ions are denoted as shown in Figure 3.
219
Methylation of microcystins. (Trimethylsilyl)diazomethane (TMSCHN2), a less hazardous
220
analogue of diazomethane with essentially the same reactivity, methylates phenols to quantitatively
221
yield the corresponding methyl ethers,14 and was therefore chosen as the reagent for O-methylation
222
of the MCs. Additionally, it was expected that TMSCHN2 would also react with the two carboxylic
223
groups present in the MCs, and therefore the reagent was preliminary tested with a standard sample
224
of MC-LR (1). Surprisingly, even with a very large excess of TMSCHN2 and long reaction times, ACS Paragon Plus Environment
12
Page 13 of 26
Environmental Science & Technology
225
only the monomethyl derivative 7 (m/z 1009.5712, C50H77N10O12, additional CH2 compared to 1) of
226
MC-LR was obtained. LC-HRMS/MS analysis of the reaction product showed that methylation
227
occurred on the Glu residue, while the MeAsp residue did not react (Figure 4). This can be
228
explained by the formation of a strong ion pair interaction between the MeAsp carboxylate and the
229
Arg guanidinium groups. The reaction was then repeated using MC-YR (3), and the bis-methylated
230
product 8 (m/z 1073.5678, C54H77N10O13, additional C2H4 compared to 3) was formed. Compound 8
231
was shown to be methylated on the Glu and Tyr residues (Figure 4), and confirmed that TMSCHN2
232
was capable of reacting with the Tyr phenolic hydroxy group.
233
Finally, the reaction was performed on two HPLC fractions of the BuOH extract (fractions
234
A and B), which respectively contained MC-HtyR (5) and MC-MhtyR (6) as the only MCs. The
235
reactions yielded, respectively, the bis-methylated products 9 (m/z 1087.5834, C55H79N10O13,
236
additional C2H4 compared to 5) and 10 (m/z 1101.5979, C56H81N10O13, additional C2H4 compared to
237
6). Unfortunately, the MS/MS spectra of 9 and 10 displayed low intensity, and most fragment peaks
238
were obscured by noise. However, the diagnostic ions formed by loss of MeAsp (ion g, m/z
239
958.5376 and C50H72N9O10+ for 9 and 972.5539 and C51H74N9O10+ for 10) were still detectable, and
240
confirmed that no methylation had occurred at this amino acid residue. These results showed that
241
both MC-HtyR (5) and MC-MhtyR (6) had a free phenolic OH. This definitely excluded the
242
presence of an O-methyltyrosine residue in both compounds, and suggested a homotyrosine-
243
containing structure as the putative structure for 5. By extension, this analysis also revealed that the
244
homotyrosine residue in 6 is likely to contain an N-methyl group.
245
Thiol Derivatization. A micro-scale derivatization technique has been recently reported, 39 which
246
allows differentiation between MCs containing the isomeric amino acids Mdha and Dhb
247
(dehydrobutryine, or 2-aminocrotonic acid), which have both been reported at position 7 of MCs.
248
The method involves the reaction of MCs with a thiol (e.g. β-mercaptoethanol) followed by LC-MS
249
analysis. When Mdha-containing MCs are present, the reaction is fast (several hours), but when
250
Dhb-containing MCs are present, complete derivatization takes several days. The presence of an ACS Paragon Plus Environment
13
Environmental Science & Technology
Page 14 of 26
251
MdhA residue in all the MCs 1-6 from the Green Lake extract was confirmed by derivatization of
252
the microcystins with β-mercaptoethanol and LC-HRMS analysis showing complete reaction after
253
12 h (see the Supporting Information section for details).
254
Ferintoic acids. The ferintoic acids are a sub-class of the anabaenopeptins, cyclic hexapeptides40
255
that are synthesized by NRPSs.41 Anabaenopeptins contain six amino acids, five of which form a
256
cyclic structure with an amide bond involving the ε-amino group on an invariant Lys residue. The
257
sixth amino acid is attached to the Lys α-amino group through a unique ureido bridge; variants of
258
the anabaenopeptins with L-Tyr, L-Arg, L-Phe, L-Ile, L-Glu, and L-Lys at this position have been
259
reported. For historical reasons, anabaenopeptins with an L-Trp residue as the exocyclic amino acid
260
are known as ferintoic acids. Two types of ferintoic acids have been described, ferintoic acid A
261
(11), and ferintoic acid B, in which the Ile residue is replaced by Val.26 Anabaenopeptins, including
262
the ferintoic acids, have been reported to inhibit protein phosphatase 142 as well as proteases such as
263
carboxypeptidase A.43
264 265
Figure 5. (a) Structures of ferintoic acids. (b) The ferintoic acid cluster from the Green Lake extract. MetO is
266
methionine sulfoxide.
267
The ferintoic acid cluster (Figure 5) contained three nodes, suggesting the presence of two
268
analogues of ferintoic acid A (11, m/z 867.4423) which we name here as ferintoic acid C (12, m/z
269
899.4143) and D (13, m/z 915.4069). The relative amounts of 11-13 (Figure 5) were estimated from
ACS Paragon Plus Environment
14
Page 15 of 26
Environmental Science & Technology
270
the areas of the relevant peaks in the extracted-ion chromatograms. The molecular formulas of 12
271
(C46H57N8O9S, one additional S atom compared ferintoic acid A) and 13 (C46H58N8O10S, one
272
additional S and O atom) did not match any known anabaenopeptin analogue.
273
The fragmentation patterns of the new compounds were compared with that of ferintoic acid A
274
(11) (Figure 6). The fragment ion at m/z 591.2941 (C31H39N6O6+) found in the MS/MS spectrum of
275
11 and due to the loss of amino acids 3 and 4, was also present in the MS/MS spectra of 12 and 13.
276
This located the difference between the three compounds to be on one of these two amino acids.
277
The ion at m/z 690.3629 (C36H48N7O7+), formed by the loss of Hty at position 4, was shifted to m/z
278
722.3352 (C36H48N7O7S+) for 12 and m/z 738.3306 (C36H48N7O8S+) for 13. Therefore, the
279
additional atoms in 12 and 13 must be located on amino acid 3. All of the other fragment ions
280
detected in the MS/MS spectra were fully consistent with this conclusion. Considering that
281
substitution of methionine (or its oxidized form methionine sulfoxide) for valine has been
282
previously observed in several anabaenopeptins, these data indicated the presence of a Met residue
283
at position 3 in ferintoic acid C (12) and a methionine sulfoxide (MetO) residue at position 3 in
284
ferintoic acid D (13).
ACS Paragon Plus Environment
15
Environmental Science & Technology
Page 16 of 26
285 286
Figure 6. High-resolution MS/MS fragmentations of ferintoic acids A, C and D. Fragments that are mass
287
shifted and therefore contain the variable amino acid are labeled in red, while the non-shifting fragments are
288
labeled in blue.
289
16S-ITS rRNA Metagenomic Analysis. The microbiome associated with the Green Lake
290
bloom was probed using a cultivation-independent approach in order to make a preliminary survey
291
of the identity of the cyanobacteria inhabiting this field sample. A 16S-ITS (internal transcribed
292
spacer) rRNA amplicon library was prepared from metagenomic DNA extracted from the collected
293
samples. PCR products were subcloned via TOPO TA cloning and five representative plasmids
294
were sequenced. All of the cyanobacterial sequences were identical at a 99% sequence identity
295
threshold, and were assigned to Microcystis aeruginosa based on the top matching sequences from
ACS Paragon Plus Environment
16
Page 17 of 26
Environmental Science & Technology
296
a BLASTn search (Table S1 in Supporting Information). Therefore, a M. aeruginosa strain (named
297
M. aeruginosa SEAGL14 in this paper) is the predominant cyanobacterial species present in the
298
2014 Green Lake bloom, and this strain is the likely producer of the MCs and ferintoic acids
299
detected in these field samples.
300
Signature sequences within adenylation domains and correlation with the synthesis of MC
301
variants. One of the most frequent structural variations in MCs is due to amino acid substitutions at
302
position 2. 44 This amino acid is selected and activated by the adenylation domain mcyBAd1.
303
McyBAd1 fragments were amplified by PCR from the metagenomic DNA using mcyB-specific
304
primers and then subcloned. Eight representative clones were sequenced that were identical at the
305
99% sequence identity threshold. The top matching sequences from a BLASTx search were
306
mcyBAd1 genes from M. aeruginosa (Table S2 in Supporting Information). Therefore, only one
307
genotype of mcyBAd1 (mcyBAd1_SEAGL14) was detected by PCR-based metagenomic screening,
308
and this was clearly associated with a Microcystis species.
309
The nine amino acids lining the putative binding pocket of the adenylation domain45 of this
310
variant of McyBAd1 were identified as DGWTIGAVKK. This signature sequence shared almost
311
complete identity with the sequence (DGWTIGAVEK) of the binding pocket of the McyBAd1
312
genotype from a number of Microcystis strains,46 which are reported to simultaneously produce
313
MCs containing Leu, Arg, or Tyr at position 2. It can be assumed that mcyBAd1_SEAGL14 should
314
have a similarly relaxed substrate selectivity, and is therefore consistent with our findings here of
315
MC variants with Leu, Phe or Tyr-related residues at position 2. It is worth noting that no MC
316
variant with an Arg residue at position 2 was detected in the Green Lake bloom extract. This is
317
presumably related to the fact that Glu331 is replaced by Lys331 in McyBAd1_SEAGL14. Glu331
318
(or Asp331) is thought to play a key role in the accommodation of basic amino acid side chains (i.e.
319
Arg);45 therefore, it can be argued that the presence of Lys331 instead of a negatively charged
320
amino acid does not allow for recognition of Arg. The Glu→Lys mutation at position 331 of
321
McyBAd1 has not been previously observed.
ACS Paragon Plus Environment
17
Environmental Science & Technology
Page 18 of 26
322 323
Table 1. Adenylation domains of the Microcystis present in the Green Lake bloom sample and other
324
reported Microcystis strains.
Microcystis sp.
SEAGL14 N-C 171/10
b
b
NIES102
Substrates
235
236
239
278
299
301
322
330
331
Leu, Phe, Tyr, (H4)Tyr, Mhty, Hty,
D
G
W
T
I
G
A
V
K
Leu, Tyr, Arg
D
G
W
T
I
G
A
V
E
Leu, Tyr, Arg
D
G
W
T
I
G
A
V
E
Leu, Arg
D
G
W
T
I
G
A
V
E
c
HUB 5-2-4-like
325 326 327
a
Binding pocket signatures identified in McyBAd1 (residue positiona)
47
As described by Conti et al. Data reported by Tooming-Klunderud et al.46 c Data reported by Mikalsen et al. 48 b
328
Signature sequences within adenylation domains and correlation with the synthesis of
329
ferintoic acids. No specific primers for adenylation domains of anabaenopeptins (and therefore of
330
ferintoic acids) are described in the literature; therefore, the metagenomic DNA from the Green
331
Lake sample was amplified by PCR using more general primers, targeting highly conserved motifs
332
in adenylation domains of NRPS. The PCR products were subcloned and six plasmids sequenced.
333
All six were related to adenylation domains of NRPSs, however, only one (ferB_Ad) shared high
334
similarity with AptB adenylation domains from the anabaenopeptin synthetase gene cluster (Table
335
S2 in Supporting Information). The AptB adenylation domain is known to recruit the amino acid
336
residue in position 3 of the anabaenopeptins. The signature sequence of the binding pocket of
337
FerB_Ad (DMWFLGGAI) shares almost complete identity with the substrate specificity code
338
(DMWFMGGVI) of AptB_Ad from the nodulapeptin synthetase cluster. 49 In nodulapeptin, the
339
AptB_Ad protein selects for Met, MetO or Ile as the substrate. Therefore, FerB_Ad is expected to
340
be responsible for the activation of Met, MetO, Ile or Val in the production of the ferintoic acids
341
from this Green Lake population.
ACS Paragon Plus Environment
18
Page 19 of 26
342
Environmental Science & Technology
DISCUSSION
343
Freshwater, marine, and terrestrial cyanobacteria produce a wide array of toxic metabolites, the
344
most common of which are the MCs.50 The presence of MCs can limit the use of lakes and
345
reservoirs for recreational activities, as drinking water reservoirs, and for irrigation. Concentrations
346
of MC-LR are highly variable in phytoplankton communities and are greatly influenced by the
347
composition of phytoplankton species present. The World Health Organisation (WHO) has set a
348
provisional guideline value for MC-LR of 1.0 µg/L in drinking water. More generally, WHO
349
established a level of 100,000 cyanobacterial cells/mL (equivalent to approximately 50 µg
350
chlorophyll-a/L if cyanobacteria dominate) as a guideline value for a moderate health alert in
351
recreational waters. 50 These provisions for adequate public warnings are the only short-term
352
measure to avoid large scale intoxications and allergies. We are evaluating the feasibility of early
353
detection of cyanobacterial blooms using similar remote and proximal sensing tools as those used to
354
monitor contamination of surface waters.51
355
Nutrient enrichment has been proved to be the most important factor responsible for cyanobacterial
356
blooms,10 even if this is a topic still opened for further studies. Similarly, our knowledge is
357
fragmentary concerning the structures and amounts of cyanotoxins produced, the fate of these
358
cyanotoxins in the food chain, and the influence of human activities on these events. Even if
359
national authorities are generally aware of the MC problem, present regulations are generally at the
360
provisional state and will need to change as knowledge increases on the structure, distribution, and
361
toxicology of MCs.
362
A general method for the rapid detection of known or novel (and potentially more toxic)
363
cyanotoxins is an urgent need to properly manage cyanobacterial blooms, and should be considered
364
in health control programs at the private, municipal, and state levels. The large structural variability
365
of cyanotoxins, and the frequent occurrence of new congeners, makes the problem of developing
366
such a method a challenging one. The Green Lake bloom is illustrative in this respect: the new MC-
367
MhtyR was by no means a minor compound, and comprised about half of the total MC content in ACS Paragon Plus Environment
19
Environmental Science & Technology
Page 20 of 26
368
the collected samples. In contrast, MC-LR, the only MC whose concentration is considered in
369
WHO guidelines, represented only about 7% of the total MCs. Similarly, the new ferintoic acids C
370
and D predominated over the known ferintoic acid A.
371
Currently, cyanotoxins are predominately monitored using reversed-phase HPLC combined
372
with MS and MS/MS. The current MS-based methods are fast and sensitive, but require the
373
availability of standards, or rely on assumptions concerning the structures of the analyzed MCs.
374
This approach can fail in the identification of new cyanotoxins. For example, the most general
375
method which can be used for the MS/MS identification of MCs by MS/MS relies on the fragment
376
peak at m/z 135 arising from side-chain fragmentation of the Adda residue. While this residue is
377
generally conserved in MCs, some examples of MCs are known in which a free OH or an OAc
378
group replaces the OMe group in Adda.52 Such MCs or others with a modification to the Adda
379
residue would not be identified by this MS/MS method.
380
This combined high-resolution LC-MS and automated molecular networking data analysis
381
approach may be used for the rapid identification of both known as well as new variants of the
382
MCs. The method does require a library of reference MS/MS spectra, but due to its inherent
383
capacity to show the similarities between structurally related molecules, can be highly effective to
384
identify even previously undescribed variants. Moreover, even though the GNPS website has only a
385
few cyanotoxins at the present time, it was highly effective in helping to identify these cyanotoxins
386
in the Green Lake bloom. It can be anticipated that future depositions to GNPS of cyanotoxin high
387
resolution MS/MS spectra will further improve the effectiveness of the method. We are currently
388
working to build such a library using both standards and the cyanotoxins present in cultured and
389
field samples.
390
In this study, we were able to determine the putative structure of the new MC-MhtyR (6) even
391
though the compound was present in the sample in insufficient quantities for NMR analysis. This
392
took advantage of a microscale methylation reaction with (trimethylsilyl)diazomethane that was
393
used here for the first time in the structural analysis of a natural product.
ACS Paragon Plus Environment
20
Page 21 of 26
Environmental Science & Technology
394
The predominant cyanobacterium in the 2014 Green Lake bloom was identified as a strain of
395
Microcystis aeruginosa. Fragments of the putative adenylation domains belonging to the
396
microcystin and ferintoic acid gene clusters were also identified. Their sequences, especially those
397
that determine substrate selectivity, matched well with the structures of the identified cyanotoxins
398
from the bloom sample. Therefore, it appears very likely that the fragments found in the
399
metagenomic DNA from the Green Lake blooms are indeed part of the microcystin and ferintoic
400
acid gene clusters.
401
ASSOCIATED CONTENT
402
Supporting Information. Tables S1 and S2 containing BLAST matches of the amplified 16S-ITS
403
rRNA, mcyBAd1, and ferBAd1 fragments. Figure S1 showing the structures of thiol derivatives of
404
MCs, and Figure S2 showing their LC-MS analysis. Figures S3-S13 showing HR-MS/MS spectra
405
of 1-8 and 11-13.
406
AUTHOR INFORMATION
407
Corresponding Author
408
* Phone: +39-081-678-504. E-mail:
[email protected] 409
Notes
410
The authors declare no competing financial interest.
411
AKNOWLEDGMENTS
412
This work was funded by the European Union’s Seventh Framework Programme (FP7) 2007–2013
413
under Grant Agreement No. 311848 (Bluegenics), and by Regione Campania under POR Campania
414
FESR 2007-2013 - O.O. 2.1 (FarmaBioNet), and partially funded by NIH R01 GM107550 to LG
415
and WHG.
ACS Paragon Plus Environment
21
Environmental Science & Technology
Page 22 of 26
416
We are grateful to Prof. Daniel Jaffe (University of Washington Seattle) and Prof. Paul Bishop
417
(University of Rhode Island) for their contribution in the sampling phase and for their
418
encouragement and support.
419
REFERENCES (1)
Pettit, G.R.; Kamano, Y.; Herald, C L.; Tuinman, A.A.; Boettner, F.E.; Kizu, H.; Schmidt,
J.M.; Baczynskyj, L.; Tomer, K.B.; Bontems, R.J. The isolation and structure of a remarkable marine animal antineoplastic constituent: dolastatin 10. J. Am. Chem. Soc. 1987, 109, 6883-6885 (2)
Schwartz, R.E.; Hirsch, C.F.; Sesin, D.F.; Flor, J.E.; Chartrain, M.; Fromtling, R.E.;
Harris, G.H.; Salvatore, M.J.; Liesch, J.M.; Yudin, K. Pharmaceuticals from cultured algae. J. Ind. Microbiol. 1990, 5, 113-123 (3)
Gerwick, W.H.; Proteau, P.J.; Nagle, D.G.; Hamel, E.; Blokhin, A.; Slate, D.L. Structure
of curacin A, a novel antimitotic, antiproliferative, and brine shrimp toxic natural product from the marine cyanobacterium Lyngbya majuscula. J. Org. Chem. 1994, 59, 1243-1245 (4)
Dixit, R.B.; Suseela, M.R. Cyanobacteria: potential candidates for drug discovery. Antonie
van Leeuwenhoek 2013, 103, 947-61. (5)
Della Sala, G.; Hochmuth, T.; Costantino, V.; Teta, R.; Gerwick, W.; Gerwick, L.; Piel, J.;
Mangoni, A. Polyketide genes in the marine sponge Plakortis simplex: a new group of monomodular type I polyketide synthases from sponge symbionts. Environ. Microbiol. Rep. 2013, 5, 809-818. (6)
Della Sala, G.; Hochmuth, T.; Teta, R.; Costantino, V.; Mangoni, A. Polyketide synthases
in the microbiome of the marine sponge Plakortis halichondrioides: a metagenomic update. Mar. Drugs 2014, 12, 5425 -5440. (7)
Luesch, H.; Moore, R.E.; Paul, V.J.; Mooberry, S.L.; Corbett, T.H. Isolation of dolastatin
10 from the marine cyanobacterium Symploca species VP642 and total stereochemistry and biological evaluation of its analogue symplostatin 1. J. Nat. Prod. 2001, 64, 907-910. (8)
Esposito, G.; Teta, R.; Miceli, R.; Ceccarelli, L.S.; Della Sala, G.; Camerlingo, R.; Irollo,
E.; Mangoni, A.; Pirozzi, G.; Costantino, V. Isolation and assessment of the in vitro anti-tumor activity of smenothiazole A and B, chlorinated thiazole-containing peptide/polyketides from the caribbean sponge, Smenospongia aurea. Mar. Drugs 2015, 13, 444-459, 16. (9)
Edwards, D.J.; Marquez, B.L.; Nogle, L.M.; McPhail, K.; Goeger, D.E.; Roberts, M.A.;
Gerwick, W.H. Structure and biosynthesis of the jamaicamides, new mixed polyketide-peptide neurotoxins from the marine cyanobacterium Lyngbya majuscula. Chem Biol. 2004, 11, 817-823.
ACS Paragon Plus Environment
22
Page 23 of 26
Environmental Science & Technology
(10) Carvalho, L.; McDonald, C.; de Hoyos, C.; Mischke, U.; Phillips, G.; Borics, G.; Poikane, S.; Skjelbred, B.; Solheim, A. L.; Van Wichelen, J.; Cardoso, A.C. Sustaining recreational quality of European lakes: minimizing the health risks from algal blooms through phosphorus control. J. Appl. Ecol. 2013, 50, 315-323 (11) Digamber Rao, B.; Babu Ramesh, M.; Ellaswamy, N.; Cyanotoxins and their potential applications. A review. Nat. Environ. Poll. Tech. 2015, 14, 203-209; Rao, P. V.; Gupta, N.; Bhaskar, A. S., Jayaraj R. Toxins and bioactive compounds from cyanobacteria and their implication on human health. J. Environ. Biol. 2002, 23, 215-224. (12) EPA-190-R-15-001,
United
States
Environmental
Agency,
Fiscal
Year
2016,
www.epa.gov/ocfo. (13) Watrous, J.; Roach, P.; Alexandrov, T.; Heath, B.S.; Yang, J.Y.; Kersten, R.D.; van der Voort, M.; Pogliano, K.; Gross, H.; Raaijmakers, J. M.; Moore, B. S.; Laskin, J.; Bandeira, N.; Dorrestein, P.C. Mass spectral molecular networking of living microbial colonies. Proc. Natl. Acad. Sci. U.S.A. 2012, 109, E1743-E1752. (14) Crowley, B.M.; Mori, Y.; McComas C.C.; Tang, D.; Boger D.L. Total Synthesis of the Ristocetin Aglycon. J. Am. Chem. Soc. 2004, 126, 4310-4317. (15) ProteoWizard Software Foundation; http://proteowizard.sourceforge.net/tools.shtml (16) UCSD Computational Mass Spectrometer Website; http://gnps.ucsd.edu/ (17) Cytoscape Software; http://www.cytoscape.org/ (18) Cytoscape Software FM3 plug-in; http://apps.cytoscape.org/apps/fm3 (19) Boutte, C.; Grubisic, S.; Balthasart, P.; Wilmotte, A. Testing of primers for the study of cyanobacterial molecular diversity by DGGE. J. Microbiol. Meth. 2006, 65, 542-550. (20) Janse, I.; Meima, M.; Kardinaal, W.E.A.; Zwart, G. High resolution differentiation of cyanobacteria by using rRNA internal transcribed spacer denaturing gradient gel electrophoresis. Appl. Environ. Microbiol. 2003, 69, 6634–6643. (21) Brett, N. A.; Dittmann, E.; Rouhiainen, L. R. Bass A., Schaub V., Sivonen K., Borner T. Nonribosomal Peptide Synthesis and Toxigenicity of Cyanobacteria. J. Bacteriol. 1999, 181, 40894097. (22) Cole, J.R.; Wang, Q.; Cardenas, E.; Fish, J.; Chai, B.; Farris, R. J.; Kulam-Syed-Mohideen, A.S.; McGarrell, D.M.; Marsh, T.; Garrity, G.M.; Tiedje, J.M. The Ribosomal Database Project: improved alignments and new tools for rRNA analysis. Nucleic Acids Res. 2009, 37 (Database issue), D141-145.
ACS Paragon Plus Environment
23
Environmental Science & Technology
Page 24 of 26
(23) Altschul, S.F.; Madden, T.L.; Schaffer, A.A.; Zhang, J.; Zhang, Z.; Miller, W.; Lipman D.J. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucl. Acids Res. 1997, 25, 3389-3402. (24) Yang, J.Y.; Sanchez, L.M.; Rath, C.M.; Liu, X.; Boudreau, P.D.; Bruns, N.; Glukhov, E.; Wodtke, A.; de Felicio, R.; Fenner, A.; Wong, W.R.; Linington, R. G.; Zhang, L.; Debonsi, H.M.; Gerwick, W.H.; Dorrestein P.C. Molecular Networking as a Dereplication Strategy. J. Nat. Prod. 2013, 76, 1686-1699. (25) Carmichael, W.W.; Beasley, V.; Bunner, D.L.; Eloff, J.N.; Falconer, I.; Gorham, P.; Harada, K.; Krishnamurthy, T.; Yu, M.J.; Moore, R.E. Naming of cyclic heptapeptide toxins of cyanobacteria (blue-green algae). Toxicon 1988, 26, 971-973. (26) Williams, D.E.; Craig, M.; Holmes, C.F.B.; Andersen, R.J. Ferintoic acids A and B, new cyclic hexapeptides from the freshwater cyanobacterium Microcystis aeruginosa. J. Nat. Prod. 1996, 59, 570-575 (27) Portmann, C.; Blom, J.F.; Gademann, K.; Juttner, F. Aerucyclamides A and B: isolation and synthesis of toxic ribosomal heterocyclic peptides from the cyanobacterium Microcystis aeruginosa PCC 7806. J. Nat. Prod. 2008, 71, 1193-1196. (28) Tillet, D.; Dittman, E.; Erhard, M.; Von Dörhen, H.; Börner, T.; Neilan, B.A. Structural organization of microcystin biosynthesis in Microcystis aeruginosa PCC7806: an integrated peptide-polyketide synthetase system. Chem. Biol. 2000, 7, 753-764. (29) Van Wagoner, R.M.; Drummond, A.K.; Wright, J.L.C. Biogenetic diversity of cyanobacterial metabolites. Adv. Appl. Microbiol. 2007, 61, 89-217. (30) Puddick, J. Spectroscopic Investigations of oligopeptides from aquatic cyanobacteria: Characterisation of new oligopeptides, development of microcystin quantification tools and investigations into microcystin production. Ph.D. Thesis, University of Waikato, Hamilton, New Zealand, 2013; http://researchcommons.waikato.ac.nz/handle/10289/7289 (31) Mackintosh, C.; Beattie, K.A.; Klumpp, S.; Cohen, P.; Codd, G.A. Cyanobacterial microcystin-LR is a potent and specific inhibitor of protein phosphatases 1 and 2A from both mammals and higher plants. FEBS Lett 1990, 264, 187-192. (32) Carmichael, W.W. Cyanobacteria secondary metabolites – the cyanotoxins. J. Appl. Bacteriol. 1992, 72, 445-459. (33) Liu, W.-T.; Ng, J.; Meluzzi, D.; Bandeira, N.; Gutierrez, M.; Simmons, T. L.; Schultz, A. W.; Linington, R. G.; Moore, B. S.; Gerwick, W. H.; Pevzner, P. A.; Dorrestein, P. C. Interpretation of Tandem Mass Spectra Obtained from Cyclic Nonribosomal Peptides. Anal. Chem. 2009, 81, 4200-4209.
ACS Paragon Plus Environment
24
Page 25 of 26
Environmental Science & Technology
(34) Mayumi, T.; Kato, H.; Imanishi, S.; Kawasaki, Y.; Hasegawa, M.; Harada, K. Structural characterization of microcystins by LC/MS/MS under ion trap conditions. J. Antibiot. 2006, 59, 710-719. (35) Benke, P. I.; Vinay Kumar, M. C. S.; Panc D.; Swarup S. A mass spectrometry-based unique fragment approach for the identification of microcystins. Analyst 2015, 140, 1198-1206. (36) Namikoshi, M.; Sun, F.; B. Choi, W.; Rinehart, K. L.; Carmichael, W. W.; Evans, W. R.; Beasley, V. R. Seven More Microcystins from Homer Lake Cells: Application of the General Method for Structure Assignment of Peptides Containing α,β-Dehydroamino Acid Unit(s). J. Org. Chem. 1995, 60, 3671-3679. (37) Sano, T.; Takagi H.; Nagano K.; Nishikawa M.; Kaya K. Accurate LC-MS analyses for microcystins using per-15N-labeled microcystins. Anal. Bioanal. Chem. 2011, 399, 2511-2516. (38) Medina, R. A.; Goeger, D. E.; Hills, P.; Mooberry, S. L.; Huang, N.; Romero, L. I.; Ortega-Barria, E.; Gerwick, W. H.; McPhail, K. L. Coibamide A, a Potent Antiproliferative Cyclic Depsipeptide from the Panamanian Marine Cyanobacterium Leptolyngbya sp. J. Am. Chem. Soc. 2008, 130, 6324-6325. (39) Miles, C. O.; Sandvik, M.; Nonga, H. E.; Rundberget, T.; Wilkins, A. L.; Rise, F.; Ballot, A. Thiol derivatization for LC-MS identification of microcystins in complex matrices. Environ. Sci. Technol. 2012, 46, 8937−8944. (40) Harada, K.; Fujii, K.; Shimada, T.; Suzuki, M.; Sano, H.; Adachi, K.; Carmichael, W.W. Two cyclic peptides, anabaenopeptins, a third group of bioactive compounds from the cyanobacterium Anabaena flos-aquae NRC 525-17. Tetrahedron Lett. 1995, 36, 1511-1514. (41) Fujii, K.; Sivonen, K.; Nakano, T.; and Harada, K.I. Structural elucidation of cyanobacterial peptides encoded by peptide synthetase gene in Anabaena species. Tetrahedron 2002, 58, 6863-6871. (42) Gkelis, S.; Lanaras, T.; Sivonen, K. The presence of microcystins and other cyanobacterial bioactive peptides in aquatic fauna collected from Greek freshwaters. Aquat. Toxicol. 2006, 78, 3241. (43) Itou, Y.; Suzuki, S.; Ishida, K.; Murakami, M. Anabaenopeptins G and H, potent carboxypeptidase A inhibitors from the cyanobacterium Oscillatoria agardhii (NIES-595). Bioorg. Med. Chem. Lett. 1999, 9, 1243-1246. (44) Kruger, T.; Christian, B.; Luckas, B.; Development of an analytical method for the unambiguous structure elucidation of cyclic peptides with special appliance for hepatotoxic desmethylated microcystins. Toxicon 2009, 54, 302-312
ACS Paragon Plus Environment
25
Environmental Science & Technology
Page 26 of 26
(45) Stachelhaus, T.; Mootz, H.D.; Marahiel, M.A. The specificity-conferring code of adenylation domains in nonribosomal peptide synthetases. Chem Biol. 1999, 6, 493-505. (46) Tooming-Klunderud, A.; Fewer, D.P.; Rohrlack, T.; et al. Evidence for positive selection acting on microcystin synthetase adenylation domains in three cyanobacterial genera. BMC Evolutionary Biology. 2008, 8, 256. (47) Conti, E.; Stachelhaus, T.; Marahiel, M.A.; Brick, P. Structural basis for the activation of phenylalanine in the nonribosomal biosynthesis of gramicidin S. EMBO 1997, 16, 4174-4183. (48) Mikalsen, B.G.; Boison, O.M.; Skulberg, J.; Fastner, W.; Davies, T.M. et al. Natural variation in the microcystin synthetase operon mcyABC and impact on microcystin production in Microcystis Strains. J. Bacteriol. 2003, 185, 2774-2785. (49) Rouhiainen, L.; Jokela, J.; Fewer, D.P.; Urmann M.; Sivonen K. Two Alternative Starter Modules for the Non-Ribosomal Biosynthesis of Specific Anabaenopeptin Variants in Anabaena (Cyanobacteria). Chem. Biol. 2010, 17, 265-273. (50) Guidelines for safe recreational water environments. Volume 1: Coastal and fresh waters; http://www.who.int/water_sanitation_health/bathing/srwe1execsum/en/index6.html# (51) Lega, M.; Kosmatka, J.; Ferrara, C.; Russo, F.; Napoli, R.M.A.; Persechino, G.; Using Advanced Aerial Platforms and Infrared Thermography to Track Environmental Contamination. Environ. Forensics (Taylor & Francis) 2012, 13, 332-338. (52) Oksanen, I.; Jokela, J.; Fewer, D. P.; Wahlsten, M.; Rikkinen, J.; Sivonen, K. Discovery of rare and highly toxic microcystins from lichen-associated cyanobacterium Nostoc sp. strain IO-102I. Appl. Environ. Microbiol. 2004, 70, 5756-5763.
ACS Paragon Plus Environment
26